The Effect of Radiation Dose to the Pharyngeal Constrictor Musculature on Weight Loss During Treatment and Feeding Tube Dependence in Head and Neck Cancer

Abstract

Radiation-associated dysphagia (RAD) is a common toxicity of radiotherapy (RT) for head and neck cancer, reported by nearly 50 percent of patients. Severe dysphagia can lead to feeding tube dependence, which significantly impacts quality of life. Studies have suggested that reduced radiation dose to the pharyngeal constrictor musculature (PCM) using intensity-modulated RT (IMRT) is associated with improved swallowing outcomes by a variety of measures. We aimed to identify dose-volume parameters of radiation to the PCM that are associated with weight loss during RT and percutaneous endoscopic gastrostomy (PEG) dependence as clinical measures of severe RAD. From a single institutional database, 115 patients with newly diagnosed head and neck squamous cell carcinoma of the oral cavity, nasopharynx, oropharynx, hypopharynx and larynx who underwent definitive or postoperative IMRT between January 1, 2014 and December 31, 2017 were identified after exclusion of patients with T1-2 stage larynx cancer, less than 6 months of follow-up and locoregional recurrence at any time. PCM was contoured prospectively and mean dose < 60 Gy was used as the dose constraint when possible. PCM dose-volume parameters were retrospectively collected from IMRT plans and logistic regression was used to analyze these data relative to percent weight loss during RT and duration of PEG use. 57% of patients had surgery prior to RT, 63% received concurrent chemotherapy and 70% had planned RT to the bilateral neck. 57% had some degree of dysphagia prior to RT. PEG tubes were placed prior to, during or just after RT in 45% of patients. The mean and median time from RT to PEG removal were 8.9 and 6.5 months respectively (range 1.9 to 36.7 months). The median percent weight loss during RT was 9.3% (range -5.7 to 23%). Definitive RT, bilateral neck volumes and concurrent chemotherapy were associated with weight loss and prolonged PEG use on univariate analysis. On multivariate analysis, female gender and concurrent chemotherapy were independently associated with prolonged PEG use and age < 60 and normal swallowing prior to RT were independently associated with weight loss > 10%. When adjusting for these significant confounding factors, PCM mean dose, max dose and V50-60 were significantly associated with prolonged PEG use: Means of these individual parameters for PEG use > 6 months were 57 Gy, 72 Gy, 80% and 54% respectively versus 48 Gy, 65 Gy, 58% and 28% for PEG use < 6 months. PCM V40 was significantly associated with > 10% weight loss during RT, with mean V40 of 85% for weight loss > 10% and 70% for weight loss < 10%. In this institutional cohort, IMRT dose to the total PCM correlated with severe RAD. Aiming to keep an overall PCM mean dose < 48 Gy and max dose < 65 Gy during IMRT planning may help reduce prolonged PEG dependence, while a PCM V40 < 70% may be associated with reduced treatment related weight loss.